The Geometry of Prompting: Unveiling Distinct Mechanisms of Task Adaptation in Language ModelsNorth American Chapter of the Association for Computational Linguistics (NAACL), 2025 |
Out-of-distribution generalization via composition: a lens through induction heads in TransformersProceedings of the National Academy of Sciences of the United States of America (PNAS), 2024 |
Wrong-of-Thought: An Integrated Reasoning Framework with
Multi-Perspective Verification and Wrong InformationConference on Empirical Methods in Natural Language Processing (EMNLP), 2024 |